1. Field of the Invention
The present invention relates to an automatic focusing circuit and more particularly, to an improvement of an automatic focusing circuit for automatically matching the focus in response to a video signal obtained from an image sensor, in a image sensing apparatus such as a video camera and an electronic still camera having an automatic focusing mechanism.
2. Description of the Prior Art
Conventionally, in an automatic focusing apparatus used in an image sensing apparatus such as a video camera and an electronic still camera, an approach utilizing a video signal itself obtained from an image sensor for evaluating the state where the focus is controlled has been developed. According to such an approach, a lot of good characteristics can be obtained. For example, there exists essentially no parallax. In addition, even if the depth of field is small and an object is located in the distance, the focus can be exactly matched. Furthermore, according to this approach, a specific sensor for automatic focusing need not be separately provided, so that the apparatus is very simple as a mechanism.
As an example of such a focus control method utilizing a video signal, a control method referred to as a so-called hill-climbing servo system has been conventionally known. The hill-climbing servo system is described in, for example, U.S. Pat. No. 4,638,364, U.S. Pat. No. 4,614,975, Japanese Patent Laying-Open Gazette No. 58505/1983 and No. 103776/1985. Briefly stated, a high frequency component of a video signal obtained from an image sensor is detected every one field as a focus evaluating value, the detected focus evaluating value is always compared with a focus evaluating value detected one field before, and the position of a focusing lens continues to be slightly vibrated so that the focus evaluating value always takes the maximal value.
FIG. 1 is a schematic block diagram showing an example of an automatic focusing circuit for a conventional video camera utilizing such a hill-climbing servo system, and FIG. 2 is a block diagram showing the details of a focus evaluating value generator 5 shown in FIG. 1.
Referring to FIGS. 1 and 2, description is made on a conventional automatic focusing circuit using a hill-climbing servo system.
Referring to FIG. 1, a video camera comprises a focusing ring 2 for moving a focusing lens 1, a focusing motor 3 for driving the focusing ring 2, and an image sensing circuit 4 including an image sensor (not shown) such as a CCD (Charge Coupled Device). The focusing lens 1 may be moved by a piezoelectric element instead of a motor. In addition, the image sensor itself (not shown) such as the CCD instead of the focusing lens may be moved.
An image formed on a surface of the image sensor by the focusing lens 1 is converted into a video signal by the image sensing circuit 4 and inputted to the focus evaluating value generating circuit 5. Referring to FIG. 2 showing the details of the focus evaluating value generating circuit 5, a luminance signal component in a video signal outputted from the image sensing circuit 4 is applied to a synchronizing separator circuit 5a and a gate circuit 5c. The synchronizing separator circuit 5a separates a vertical synchronizing signal VD and a horizontal synchronizing signal HD from the inputted luminance signal and applies the same to a gate control circuit 5b. The gate control circuit 5b sets a rectangular sampling area in a central portion of a picture in response to the inputted vertical synchronizing signal VD and horizontal synchronizing signal HD and a fixed output of an oscillator (not shown). The gate control circuit 5b applies a signal for opening or closing a gate every field to the gate circuit 5c so that passage of the luminance signal is permitted only in the range of the sampling area. The gate circuit 5c may be provided anywhere in the former stage of an integration circuit 5f as described below.
Only the luminance signal corresponding to the range of the sampling area is applied to a high-pass filter 5d every field by the gate circuit 5c . The high frequency component of the video signal separated by the high-pass filter 5d is amplitude-detected by a detector 5e, the detected output being applied to the integration circuit 5f. The integration circuit 5f integrates every filed the detected output applied thereto, the integrated output being applied to an A/D converter 5g. The A/D converter 5g converts the integrated value inputted thereto into a digital value and supplies the digital value as a focus evaluating value in the current field. The supplied focus evaluating value is applied to a first memory 6 as described below.
Returning to FIG. 1, a focus evaluating value outputted from the focus evaluating value generating circuit 5 is stored in the first memory 6. When a focus evaluating value in the next field is then outputted from the focus evaluating value generating circuit 5, data stored in the first memory 6 is transferred to a second memory 7. More specifically, the contents of the first memory 6 and the second memory 7 are updated every field, so that the newest focus evaluating value and a focus evaluating value one field before are always stored in the first memory 6 and the second memory 7, respectively. The contents of the two memories 6 and 7 are inputted to a comparator 8 and compared therein. The compared output is applied to a focusing motor control circuit 9.
As a result of comparison by the comparator 8, when the focus evaluating value stored in the first memory 6 is larger than that stored in the second memory 7, the focus evaluating value is increasing, so that the focusing motor control circuit 9 maintains the current rotational direction of the focusing motor 3 in response to an output of the comparator 8. On the other hand, when the focus evaluating value stored in the first memory 6 is smaller than that stored in the second memory 7, the focus evaluating value is decreasing, so that the focusing motor control circuit 9 reverses the rotational direction of the focusing motor 3 in response to the output of the comparator 8. The focusing ring 2 supporting the focusing lens 1 continues to move in the direction of increasing a focus evaluating value by such movement of the focusing motor 3, so that an in-focus state is achieved. After achieving the in-focus state, the focusing ring 2 and the focusing lens 1 continue to be vibrated back and forth in the vicinity of the maximal point of the focus evaluating value.
In the above described hill-climbing servo system, if only the slope of a focus evaluating value is detected, the lens 1 is not stopped in the defocused position by driving the focusing lens 1 in the direction of always increasing the focus evaluating value even if an object is changed, so that very good follow-up characteristics can be achieved.
However, such a hill-climbing servo system suffers from the following significant disadvantage caused by continuing to vibrate the position of the focusing lens.
A first disadvantage is that since the focusing lens is not stopped even in the in-focus state, a picture continues to be vibrated even if an object at rest is in focus. For example, the focal length of a lens currently used in a television camera is changed by rotating the focusing ring, so that the angle of field of a sensed image is changed. Therefore, in the above described system in which the focusing ring continues to be vibrated even in the in-focus state, the object on the picture becomes large or small with a particular period, resulting in a very unclear picture.
A second disadvantage is directed to a power consumption. There are many cases where a home video camera currently utilizes a battery as a power supply due to the portability thereof. When a focusing motor is always driven as in the above described hill-climbing servo system so that the forward rotation and the reverse rotation are repeated, more power is consumed, as compared with when the focusing motor is rotated in a constant direction, due to rash current, so that the time period during which an image can be recorded by using such a battery becomes short.
Additionally, since the focusing ring is always rotated, a problem of wear of a gear occurs, for example.
In order to overcome these disadvantages, there is proposed a system for detecting the maximal point where a focus evaluating value is changed from an increasing tendency to a decreasing tendency by driving a focusing ring in a one-way direction, and returning the focusing ring to the maximal point and stopping the same therein, which is disclosed in Japanese Utility Model Laying-Open Gazette No. 135712/1985. In detecting the maximal value, focus evaluating values are compared every one field, the larger focus evaluating value is always stored as the maximum value, and the maximum value is determined as the maximal value when it is determined that the current focus evaluating value has dropped, by a predetermined threshold value, from the maximum value.
On the other hand, in a video camera, the position of the focus must be changed, following an object which changes momentarily. Even after the lens is once stopped in the in-focus position as described above, hill-climbing operation of the lens must be resumed when the distance between the object and the lens is changed.
Therefore, an approach of determining that an object changed when the focus evaluating value changed, by more than a predetermined threshold value, while the focusing lens is stopped and resuming hill-climbing operation is proposed by one of the inventors of the present invention, which is disclosed in Japanese Patent Application No. 252545 filed Nov. 11, 1985. According to this approach, the position of the focus can be changed following an object which changes momentarily. On the other hand, this approach suffers from two disadvantages as described below.
It is assumed that an object moves vigorously during hill-climbing operation. In such a case, the focus evaluating value is increased or decreased as the whole in such a manner that increase or decrease of the focus evaluating value concerning the matching of the focus caused by displacement of the focusing lens itself is overlapped with increase or decrease of the focus evaluating value caused by unintentional movement of the hand and movement of an object.
If and when such unintentional movement of the hands and movement of the object are too vigorous, the effect on the focus evaluating value caused thereby becomes dominant. The point where the focus evaluating value is changed from an increasing tendency to a decreasing tendency by unintentional movement of the hands and movement of an object is recognized as the maximal point corresponding to the in-focus position although the focus is not actually matched, so that the focusing lens may be stopped therein. After the lens is stopped in the erroneous maximal point, hill-climbing operation is resumed if the unintentional movement of the hands and the movement of the object are continued, so that extra time may be somewhat required until the focus evaluating value attains the in-focus position. On the contrary, if the unintentional movement of the hands and the movement of the object are stopped and the focus evaluating value is not changed thereafter, the hill-climbing operation is not resumed, so that the lens continues to be stopped in the defocused position.
A second disadvantage is that a threshold value for resuming hill-climbing operation must be decreased to improve the follow-up characteristics of the position of the focus. However, when the threshold value is decreased, the hill-climbing operation is resumed due to slight movement of an object even in the in-focus state, so that the lens moves, resulting in an unclear picture. Therefore, if the threshold value is set large, the possibility that the lens continues to be erroneously stopped in the defocused position is increased.
In order to overcome these two disadvantages, a technique of providing defocused state detecting means for detecting the presence or absence of a particular high frequency component in a video signal and resuming hill-climbing operation when the defocused state is detected is disclosed in Japanese Patent Laying-Open No. 86972/1985.
However, it is difficult, for the following reasons, to exactly determine whether the focus is matched or not depending on the presence or absence of a particular high frequency component. More specifically, there is a variety of distributions of the spacial frequency of an object itself. There usually exist an object including few particular frequency components even when it is in the most suitable in-focus state and an object including sufficiently particular frequency components even when it is not in the most suitable in-focus state. Therefore, in the former object, it is determined that the focus is not matched even if the object is in the most suitable in-focus state, so that a lens is not stopped. On the other hand, in the latter object, if the lens is stopped while the object is in the defocused state to some extent, it is determined that the focus is matched, so that the defocused state is maintained.
Furthermore, when an object is dark, the S/N ratio in the high frequency of a video signal is deteriorated as compared with that in the low frequency thereof, so that it is likely that it is determined that the focus is matched due to the presence of a noise component in the high frequency. In order to avoid this, when the threshold value for determining the presence or absence of a particular high frequency component is increased, there are many cases where the lens is not stopped even if the object is in focus.
As described in the foregoing, in a conventional automatic focusing circuit using a hill-climbing servo system, a lens continues to be vibrated in the in-focus state, resulting in an unclear picture. In addition, if the vibration would be prevented, malfunction that the lens is stopped in the defocused position occurs.
Meanwhile, some video cameras and electronic still camera have both the automatic focusing mechanism and a zoom mechanism. For example, in the zoom mechanism of the video camera, it is known that the depth of field depends on the zoom region, that is, the place where a zoom lens which is a variable-power lens is located between the telephoto position and the wide-angle position. More specifically, if the zoom is located in the wide-angle region, the depth of field becomes large. This means that the in-focus state is easily maintained even if an object slightly moves back and forth from the in-focus position. On the other hand, if the zoom is located in the telephoto region, the depth of field becomes small. This means that it is difficult to maintain the in-focus state.
Thus, if and when the automatic focusing mechanism to fix a focusing lens in the maximal point of the focus evaluating value as described above is used with the above described zoom mechanism, the amount of change of the focus evaluating value is smaller if the zoom is located in the wide-angle region and larger if the zoom is located in the telephoto region, with respect to a predetermined amount of displacement of the focusing lens. Thus, at the time of the above described automatic focusing operation, if a threshold value for determining the maximal point and a threshold value for resuming hill-climbing operation are fixed values, more time is required for detecting the threshold values when the zoom is located in the wide-angle region, as compared with the situation where the zoom is located in the telephoto region, so that it is difficult to perform fast automatic focusing operation.